Assignment:
SPSS t-Tests (Independent Samples and Paired Samples)
Scenario:
The color-distance illusion (Kantowitz, Roediger, & Elmes, 1997) suggests that in art, warm colors (yellow, orange, red, and blends of these colors) appear to move toward the viewer when viewed from a distance, while cool colors (blue, green) seem to recede or move further away from the viewer when viewed from a distance. Suppose a researcher wanted to determine if it is true. The researcher could set up the following experimental design to test this:
Suspend one painting completed with red, yellow and/or orange WARM colors from the ceiling near the far end of a long room. Have participants stand against the wall at the opposite end of the room and make judgments regarding the distance of the painting from them. Suspend a second painting completed with blue and/or green COOL colors from the same area of the ceiling near the far end of the same long room. Have participants stand against the wall at the opposite end of the room and make judgments regarding the distance of the painting from them.
Ways in which this experiment would be controlled: same long room would be used (30 feet in length) and the painting would hang from same location in room (24 feet from the opposite wall). Also, participants would stand in same spot in room from which to make distance judgments (24 feet from the painting).
Your Tasks:
The data for this experiment can be carried out and analyzed with either a Between-Subjects Design or a Within-Subjects Design.
I. Use the following measures to set up both data sets - one data set using a Between-Subjects set-up (hint - will have 20 participants), and one data set using a Within-Subjects set-up (hint - will have 10 participants):
When participants judged the distance of the WARM paintings, they gave (from 1st to last participant) the following judgments of distance (in feet):
10, 15, 19, 16, 20, 11, 14, 9, 13, 19
When participants judged the distance of the COOL paintings, they gave (from 1st to last participant) the following judgments of distance (in feet):
30, 22, 25, 23, 27, 29, 21, 24, 24, 27
More hints:
For the Between-Subjects dataset:
You will set up one column for the Independent Variable (e.g., "type of painting") and then identify the different levels of this Independent Variable under the "variable view" screen as "values" (e.g., "warm" and "cool").
Next, you will set up one column for the Dependent Variable (e.g., "distance"). Then, for each participant, and you'll enter the value of the dependent variable attributable to the participant in the Dependent Variable column.
For the Within-Subjects dataset:
You will set up one column for each Independent Variable LEVEL (e.g., "warm" and "cool") that your participant is experiencing. Then, you will enter the DEPENDENT variable data (their ratings) for each participant in the appropriate LEVEL column.
*If you are having trouble entering data, REFER TO YOUR LECTURE NOTES on Independent & Paired Samples T-Tests*
II. Run the appropriate t-Tests (either a Paired-Samples t-Test or an Independent Samples t-Test under "Analyze" ? "Compare Means") based on the set-up used for each data set. Remember, you will be running BOTH t-Tests.
III. Print out the output for each analysis and indicate where to locate the response to each question as well as filling in below.
1.) The means for each level of the Independent Variable
2.) The standard deviation of each level of the Independent Variable
3.) The t values and degrees of freedoms for the t-test performed for each data set
4.) What is the p-value or significance level calculated for each test and is that p-value significant or non-significant?
IV. Create a BOXPLOT for the Between-Subjects design (Hint: variable - distance and category axis - type of painting painting) and attach. Please label values for full credit
V. Report your findings for each t-test below. Make sure to include if the t-test was significant and at what level (p-value). Please refer your answer back to the original hypothesis.
Write your response for the Between-Subjects analysis using your APA format handout and be sure to include your supporting statistics :
Write your response for the Within-Subjects analysis using your APA format handout and be sure to include your supporting statistics :
VI. Interpretation of SPSS data.
Scenario: A teacher was interested in knowing which students preferred different snack options and if the snack options had a relationship with favorite activity. So, the teacher asked the students what their favorite snacks were. She separated the snacks into three groups (fruit, granola bars, and chocolate). She then asked students what their favorite activity was. She separated the activities into four groups (playing sports, watching TV, playing video games, and reading a book).
1) The teacher conducted a Between-Subjects t-test using snack option as the independent variable and activity as the dependent variable. She decided to look at students who ate either chocolate or fruit. Here is the output.
Independent Samples Test
|
|
Levene's Test for
Equality of Variances
|
t-test for Equality of Means
|
|
|
|
|
|
|
|
|
|
95% Confidence
Interva of the
|
|
|
|
|
|
|
|
Mean
|
Std. Error
|
Difference
|
|
|
F
|
Sip.
|
t
|
df
|
Sip. (2-tailed)
|
Difference
|
Difference
|
Lower
|
Upper
|
|
Activity Equal variances
assumed
|
2.218
|
.153
|
C C
|
19
|
.528
|
-.31818
|
.49246
|
-1.34892
|
.71255
|
|
Equal variances
|
|
|
|
|
|
|
|
|
|
not assumed
|
|
|
18.586
|
.521
|
-.31818
|
.48641
|
-1.33778
|
.70142
|
What is the t-test value ?
What is the significance level ?
What do the results of this t-test indicate? In your explanation be sure to include the rationale behind your decision and the statistics to support it. This should look just like your APA formatting handout!
2) Second, the teacher conducted a t-test of independent samples to test if there was a difference between the activities students choose based on students who reported their favorite snack was fruit compared to those who said their favorite snack was a granola bar. Here is the output.
Independent Samples Test
|
|
Levene's Test for
Equality of Variances
|
t-test for Equality of Means
|
|
|
|
|
|
|
|
|
|
95% Confidence
Interva of the
|
|
|
|
|
|
|
|
Mean
|
Std. Error
|
Difference
|
|
|
F
|
Sig.
|
t
|
df
|
Sig. (2-tailed)
|
Difference
|
Difference
|
Lower
|
Upper
|
|
Activity Equal variances
|
|
|
|
|
|
|
|
|
|
|
assumed
|
17.009
|
.001
|
-2.814
|
18
|
.011
|
-.88889
|
.31585
|
-1 55247
|
-.22531
|
|
|
|
Equal variances
|
|
|
|
|
|
|
|
|
|
|
not assumed
|
|
|
-2.530
|
8.000
|
.035
|
-.88889
|
.35136
|
-1.69914
|
-.07884
|
What is the t-test value ?
What is the significance level?
What do the results of this t-test indicate? In your explanation be sure to include the rationale behind your decision and the statistics to support it.